HYDROLOGY AND WATER RESOURCES IN ARIZONA AND THE SOUTHWEST, VOLUME 7, p.
241 -248.
Proceedings of the 1977 meetings of the Arizona Section of the American Water Resources Association
and the Hydrology Section of the Arizona Academy of Science, held in Las Vegas, Nevada, April 15 -16.
ROOT SYSTEM OF SHRUB LIVE OAK IN RELATION TO
WATER YIELD BY CHAPARRAL
by
Edwin A. Davis
ABSTRACT
The root system of shrub live oak (Quercus turbinella) was studied in an initial
effort to classify the major Arizona chaparral shrubs as potential users of soil water
based on root system characteristics. The root system was of the generalized type with
a taproot, many deeply penetrating roots, and a strong lateral root system.
Roots
penetrated 21 feet to bedrock through cracks and fractures in the rocky regolith.
A
dense network of small surface laterals radiated from the root crown and permeated the
upper foot of soil.
Because of its root system, shrub live oak is well adapted to
utilize both ephemeral surface soil moisture as well as deeply stored moisture.
Emphasis is placed on the importance of a knowledge of the root systems of chaparral
shrubs and depth of the regolith in planning vegetation conversions to increase water
yield.
INTRODUCTION
Research in Arizona has shown that, under certain conditions, water yield from
experimental chaparral watersheds can be increased by converting from brush to grass
(Hibbert, et al. 1974).
Similar vegetation conversions in California chaparral have
also resulted in increased runoff (Rowe and Reimann 1961, Hill and Rice 1963).
The
usual explanation for the results of such conversion is that the deep- rooted brush is
able to obtain more water from deeper soil layers than the shallow- rooted grasses, thus
reducing the amount available for water yield.
Dense stands of chaparral frequently consist of a mixture of shrub species.
Increased water yields obtained by converting these areas to grass must, therefore, be
attributed to the average effect of controlling a mixture of brush species.
Water use
differences among species go undetected.
Knowledge of the root habits of chaparral shrubs and depth of the regolith occupied by chaparral is prerequisite to improved treatments for increasing water, wildlife,
grazing, and recreational values of chaparral.
Such knowledge is also basic to problems of plant establishment, adaptation, distribution, ecesis, competition, drought
tolerance, soil stability, and succession following fire, herbicide treatments, or root
plowing.
In brush control, the relationship between rooting depth and movement and
persistence of herbicides in the soil may help to explain species response differences
and aid in the development of improved brush control methods.
Since the early work of Cannon (1911) with desert plants near Tucson, Arizona,
little information has been added to our knowledge of the root habits of native woody
plants in Arizona.
Phillips (1963) reported finding what appeared to be living mesquite
roots 175 feet below the original ground surface of an open pit mine in Arizona.
Except for limited observations on shrub live oak (Quercus turbinella Greene) very little
is known about the root systems of chaparral shrubs of Arizona.
This did not limit
research in water yield improvement since yield comparisons were made between mixed
chaparral and grass.
But a lack of knowledge concerning root systems and water use of
the dominant chaparral shrubs could limit the scope of future research and management
projects concerning water yield and wildlife habitat.
Shrub live oak is the dominant shrub species in much of Arizona chaparral.
The
root system of a typical plant was thoroughly studied by means of a quantitative determination of root mass distribution vertically and laterally, followed by further excavation and examination of a central section of the root system.
Results are then interpreted as they relate to watershed management practices.
The author is Plant Physiologist, USDA Forest Service, Rocky Mountain Forest and Range
Experiment Station, Forestry Sciences Laboratory located at Tempe, Arizona, in cooperation with Arizona State University; central headquarters are maintained at Fort Collins,
in cooperation with Colorado State University.
241
-STUDY AREA
The shrub live oak bush was excavated in central Arizona on the Three Bar Wildlife Area, at 3300 -feet elevation on the easterly face of the Mazatzal Mountain Range
west of Lake Roosevelt.
Chaparral at the study area consisted of a mixture of
s clerophyllous woody shrubs dominated by shrub live oak. Other species included
birchleaf mountainmahogany (Cercocarpus betuloides Nutt.), yellowleaf silktassel
(Garrya flavescens S. Wats.), sugar sumac (Rhus ovata S. Wats.), and hollyleaf buckthorn
(Rhamnus crocea Nutt.).
These shrubs comprisé a postfire complex of sprouting and non sprouting s ru s that regrew following a wildfire in 1959.
The climate of chaparral zones in Arizona has a biseasonal precipitation pattern
characterized by summer rainfall, fall drought, winter precipitation and spring drought.
Annual precipitation at the study area averaged 25.7 inches over the past 18 years, with
about 61% occurring as rain or snow from November through April.
Soil at the root excavation pit was typical of that on nearby experimental waterIt is classified in the Barkerville series.
The soil is a very gravelly sandy
loam derived from granitic parent material; it is slightly acid and has an A, C, R,
horizon sequence.
It is well drained and has infrequent surface runoff. Permeability
is moderately rapid.
Seismic exploration on the watersheds indicated that the coarse grained granite is weathered and fractured 20 -40 feet deep. The presence of such a
deep regolith has significant hydrologic and ecologic implications.
sheds.
The excavated bush, located on a 40% east -facing slope, was a vigorous, mature
specimen typical of shrub live oak in the area.
Charred snags indicated that it had
burned in the 1959 wildfire; judging from the size and appearance of the root crown it
also may have burned in previous wildfires.
The 8 foot -tall bush had canopy dimensions
of 18.5 x 19.0 feet.
Its canopy consisted of 12- year -old regrowth stems which sprouted
from the root crown following the wildfire.
Sprouts came from an almost circular area
5
feet in diameter.
METHODS
The first phase of the study consisted of a quantitative determination of root
The trench
A trench was cut along the north side of the bush with a bulldozer.
was 11 feet deep at the upslope end, 21 feet long, and 19 feet wide, with the bottom
nearly level. The face exposed for sampling was about 3 feet out from the edge of the
bush's root crown.
Samples were taken to determine root mass distribution in four,
1- foot -thick transections (slices) of the regolith starting at the trench face and
moving in toward the bush.
Pattern of root distribution was determined by removing
1- cubic -foot blocks from each transection by means of picks and a pneumatic chipping
hammer, separating roots from soil by passing the soil through a 1/4 -inch mesh sieve,
and subsequently weighing the ovendried roots. The two inner transections included
In
parts of the root crown, but its weight was not included with the weight of roots.
constructing lateral and vertical distribution patterns only data to the 6 -foot depth
were used, because the number of lateral blocks varied below this depth.
mass.
In the second phase of the study, a backhoe was used to dig a 6- foot -wide pit in
the bottom of the trench. At the upslope end of the pit, at a depth of approximately
20 -feet, hard rock was reached that could not be broken.
Roots broken during digging
were identified with those protruding from the face of the pit, tagged, and later reconnected.
The root system in a 1 -2 foot transection of soil beneath the center of the
bush was then exposed by careful excavation and reconstruction.
Roots in this excavation were allowed to hang against the face of the pit.
Next, a portion of the surface
lateral root system was exposed in a semicircular trench, 18 inches wide and 1 foot deep,
about 2 feet out from the base of the bush.
Exposed roots were painted white with vinyl
latex paint for photographic documentation.
RESULTS
QUANTITATIVE RESULTS
Roots were found in every cubic foot of soil in spite of the rocky regolith.
Root
weights per cubic foot block varied from less than 1 g to 2804 g.
Because of the hardness of the regolith, many roots penetrated crevices, fractures, and seams of weathered
material.
The greatest accumulation of root mass was centered beneath the root crown,
the zone occupied by the taproot.
Distribution of root mass in the other transections
followed the same trend; root mass decreased, in general, with lateral and vertical
distance from the root crown.
Lateral distribution of root mass was determined from the two inner transections
of the regolith to a depth of 6 feet. Each transection was made up of 66, 1- cubic -foot
blocks.
Weight of roots at each lateral location was based on 12, 1- cubic -foot blocks.
242
Forty -seven percent of the
beneath the root crown; 73%
Although there was a marked
column of the transection,
due to the lack of uniform
of large roots.
root mass was located in a 4- square -foot (2 x 2) column
was located within an 8- square -foot (2 x 4) central column.
decrease in root mass beyond the central 4- square -foot
root mass did not steadily decline with distance downslope
penetrability of the regolith and the irregular distribution
Vertical distribution of root mass to a depth of 6 feet was determined by combining
Weight of roots for each depth was
data for the four transections according to depth.
based on 30, 1- cubic -foot blocks of soil. Vertical distribution of root weight from the
26, 29, 18, 12, 9, and 6
1
-foot
increments
was
as
follows:
soil surface downward in
The greatest accumulation of root mass was in the top 2 feet of soil. Below
percent.
feet there was a gradual decrease in root mass with depth.
2
A gross estimate of the root /top ratio of the excavated bush was made by estimating
total weight of the root system based on 63.8 pounds ovendry roots in the 239 cubic feet
Assuming that the weighed roots represented 1/3 of the entire root
of soil sampled.
system, which is a probable maximum, then the total ovendry weight of roots was 191.4
Ovendry weight of the clipped top at the start of
pounds, excluding the root crown.
Thus, at this stage in the life of the plant, the weight
excavation was 101.6 pounds.
If
of the root system exceeded that of the top; the estimated root /top ratio was 1.9.
the massive root crown is included in the weight of the root system, then its estimated
weight is 321.8 pounds, and the root /top ratio would be 3.2.
DESCRIPTION OF THE ROOT SYSTEM
Following the quantitative phase of the study, a central portion of the root
system beneath the root crown of the bush was excavated and displayed to nearly its
The taproot had already been removed for weighing in the
full extent (Figure 1).
quantitative transections. It consisted of a network of main roots that fused into a
The taproot was attached to the
single, rapidly tapering root with numerous branches.
base of a massive root crown; some of its branch roots penetrated 21 feet to bedrock.
A long surface
The horizontal spread of vertically penetrating roots was 16 feet.
This root
lateral root extended downslope 22.7 feet before turning sharply downward.
A lateral growing diagonally
was only partially excavated when figure 1 was taken.
Downslope
main
laterals
were more
upslope extended 11 feet before it turned downward.
abundant than upslope laterals.
A dense network of fine surface laterals that radiated from the root crown comAbundance of these roots was
pletely permeated the upper foot of soil (Figure 2).
Diameters of
greater from the downslope half of the bush than from the upslope half.
roots excavated in the surface trench ranged from 1 mm or less up to 1 cm; the majority
of roots were less than 2 mm in diameter.
A unique feature of this particular root system was a long, thin, dense network
of intertwining roots that extended downward as a vertical plate, oriented crosswise
The plate was about 10 feet
to the direction of the slope of the hillside (Figure 3).
long, 3.5 feet wide, and 3 inches thick; it was extremely dense, especially near the
At 8 -10 feet some of the larger roots were flattened into irregularly
soil surface.
shaped belt -like strips by their confined growth along fracture planes in the rock.
These roots were 2 -5 cm wide and 0.4 -1.0 cm thick. The plate followed a narrow seam
of highly weathered granite and sandy clay that penetrated the rocky regolith. Root
Moisture, nutrient, and aeration
grafts were common where roots were rigidly confined.
conditions in the seam were apparently ideal for extensive root development.
The presence of bedrock at approximately 21 feet prevented further extensive root
penetration; roots were forced to turn in a more or less horizontal direction and grow
Diameters of these terminal roots were
along a gently sloping surface of the bedrock.
A few roots penetrated small cracks in the bedrock and may have ex1.5 to 5.0 mm.
Unsuberized fine roots were found growing in thin layers of
tended a few feet deeper.
soil between plates of rock at the 20- to 21 -foot depth.
DISCUSSION
Its
Shrub live oak is well adapted to the chaparral environment in Arizona.
vigorous capacity to sprout from the root crown allows it to thrive under a natural
The root crown gradually enlarges and the number of
regimen of periodic wildfires.
Pond and Cable (1960) found that there
stems increases following top removal by fire.
Even
were nearly six times as many live stems a year after burning than originally.
after four annual burns the number of stems was still greater than pretreatment counts.
This prodigious regenerative capacity is supported by stored energy reserves in the
root crown and roots, and by supplies of water and nutrients made available by an extensive root system.
243
Figure 1.
Midsection of the root system of shrub live
oak to a depth of 21 feet. After the
quantitative phase of the study a central
portion of the root system beneath the root
crown of the bush was excavated and dis-
played to nearly its full extent.
244
Figure 2.
Downslope side of the surface lateral root system showing the dense
layer of small roots radiating from the root crown, and some
shallow main laterals.
Figure 3.
Roots penetrated fractures in the rocky
regolith forming plates of roots along
fracture planes. Some roots were flattened
into irregularly shaped belt -like strips.
Numerous self -grafts occurred where roots
grew in the rigid confines of cracks and
fractures.
246
In accordance with the terminology used for classifying root systems (Cannon 1911,
1949; Weaver and Clements 1938) shrub live oak is of the generalized type in which both
Its extensive, deeply penetrating root system
taproot and laterals are well developed.
Thus, it can
allows it to draw upon deep soil moisture during periods of drought.
maintain an active physiological state for a longer period of time than less extensively
rooted plants.
Shrub live oak also has a highly developed surface root system to
A dense layer of surface roots gives it a
utilize ephemeral surface soil moisture.
competitive advantage over grasses, and may explain the difficulty in establishing and
maintaining perennial grasses in stands of burned shrub live oak in which regrowth is
not suppressed.
Dominance of the oak in stands of mixed chaparral may also be explicable
in terms of competitive advantages associated with its extensive root system.
Self -grafts of shrub live oak roots were common. ..Grafting occurred when roots grew
Saunier andWagle (1965) observed self in the restrictive confines of fractures.
Of several California
grafts of shrub live oak as well as intraspecific grafts,
chaparral shrubs observed by Hellmers et al. (1955), Quercus dumosa (California scrub
oak), a close relative of shrub live oak, was the only speciess in wwhich root -grafting
was observed.
Root grafts have also been reported for several other oak species
(Graham and Bormann 1966).
Most of the observed roots of the excavated oak, including the surface laterals,
were suberized.
Since the plant responds rapidly to soil moisture following drought
and to soil -applied herbicides following rain, it is possible that the suberized roots
Direct measurements by Chung and Kramer (1975) clearly
are active absorbers of water.
indicate that suberized roots usually constitute the major absorbing system of woody
plants.
The wide lateral spread of surface laterals has important hydrologic and ecologic
What may be a fairly open stand of shrub live oak above ground, dependimplications.
ing on the length of time following a wildfire, may actually be a relatively closed
one from the standpoint of overlapping root systems.
Final root pattern and growth potential are genetically determined when soil
factors are favorable but may be profoundly modified by soil structure, water content,
Roots of shrub live oak are
and aeration (Weaver and Clements 1938, Pearson 1974).
capable of penetrating deeply either in a highly weathered regolith or in fractured and
When
downward
growth
is prevented by an impenetrajointed rock underlying the solum.
ble layer, however, the roots are necessarily restricted to the upper soil layer.
Shrub live oak bushes have been observed growing in 6 to 10 inches of soil overlying
highly indurated sediments (Saunier and Wagle 1967).
This indicates that its distribution is not restricted to deep soils or weathered and fractured subsoils.
Quercus dumosa was reported by Hellmers et al. (1955) to have the deepest root
In roadcut observations
penetration o1 fh-e California chaparral shrubs they observed.
its roots were found penetrating fractured rock to a depth of 28 feet below ground
surface.
Unlike the shrub live oak in this study, R. dumosa had few feeder roots in
the top 6 inches of soil.
Present vegetation manipulation practices for the conversion from brush to grass
for water yield improvement are not intentionally selective from the standpoint of
Some selective hand applications of pelleted herbicides have been made
brush species.
Since difin which desirable browse species were not treated (Davis and Pase 1969).
ferences in water use by chaparral shrubs have not been defined, treatment selectivity
The
deeper
and
more
extensive
a
plant's
root
on this basis has not been possible.
In
system the longer the plant can remain active and transpire during dry seasons.
well- aerated soils deep- rooted species can remove water from considerable depths.
Deep- rooted shrubs are potentially much heavier water users than shallow- rooted species
Although the greatest differences in
in areas with deep soils and limited rainfall.
rooting depth occur between grasses and trees or shrubs, considerable variation may
Possible differences among chaparral shrubs may provide a
occur within each group.
partial basis for identifying low and high water users.
The sprouting ability of a shrub can have a bearing on the extent of its root
Sprouting species have the potential for developing deeper and wider spreading
In areas that have wildfires, sprouting species are
root systems than nonsprouters.
Consequently, they
older than nonsprouters since they survive the effects of fire.
Differences in root systems may occur among
should have more extensive root systems.
Plants in arid and semiarid regions are
sprouting species due to hereditary factors.
adapted by virtue of a variety of morphological, anatomical, and physiological characteristics, and may or may not be deep rooted.
Many desert plants do not have deep root
systems (Barbour 1973), but have other attributes that aid their survival (Gindel 1973).
It is possible, therefore, that differences exist in the root habits of Arizona chaparral shrubs.
Striking differences in root systems were found to occur among California chaparral species (Hellmers, et al. 1955).
Differences were also noted in the
abilities of various species to penetrate rock crevices.
system.
247
A more detailed knowledge of characteristics of the major Arizona chaparral shrubs
would allow for more effective management for multiple use objectives, assuming the
availability of a suitable, registered brush control herbicide for chaparral conversions.
Needed is a classification of chaparral shrubs which includes water -use characteristics,
type and depth of root system, susceptibility to single and repeated burns, response to
herbicides, sprouting ability, and browse value.
In some of these categories considerable information is already available; but for others there is little or no information.
This study of shrub live oak represents an initial effort to increase our knowledge of the root systems of chaparral shrubs.
REFERENCES CITED
Barbour, M. G.
1973.
Desert dogma reexamined:
spacing.
Am. Midl. Nat.
89:41 -57.
root /shoot productivity and plant
Cannon, W. A. 1911.
Publ. No. 131.
The root habits of desert plants.
Cannon, W.
A tentative classification of root systems.
1949.
A.
Carnegie Inst. Washington
Ecology 30:542 -548.
Chung, H., and P. J. Kramer.
1975.
Absorption of water and 32P through suberized and
unsuberized roots of loblolly pine.
Can. J. For. Res. 5:229 -235.
Davis,
E.
A. and C. P. Pase.
1969.
Selective control of brush on chaparral watersheds
with soil- applied fenuron and picloram.
U. S. For. Serv. Res. Note RM -140.
4 p.
Rocky Mt. For, and Range Exp. Stn., Fort Collins, Colo.
Gindel, I. A. 1973.
A new ecophysiological approach to forest -water relationships in
arid climates.
Dr. W. Junk B. V., Publishers, The Hague.
Printed by Pitman
Press, Great Britain.
142 p.
Graham, B.
F.
and F.
H.
Bormann.
Natural root grafts.
1966.
Bot. Rev. 32:255 -292.
Hellmers, H., J. S. Horton, G. Juhren, and J. O'Keefe.
1955.
Root systems of some
chaparral plants in southern California.
Ecology 36:667 -678.
Hibbert, A.R., E. A. Davis, and D. G. Scholl.
1974.
Chaparral conversion potential
in Arizona.
Part I:
water yield response and effects on other resources.
USDA
For. Serv. Res. Pap. RM -126.
36 p.
Rocky Mt. For. and Range Exp. Stn., Fort
Collins, Colo.
Hill,
L.
W., and R.
M.
Rice.
1963.
yield in southern California.
Converting from brush to grass increases water
Range Manage.
16:300 -305.
J.
Pearson, R. W.
1974.
Significance of rooting pattern to crop production and some
problems of root research, p. 247 -270.
In W. E. Carson (ed.)
The plant root and
its environment.
The Univ. Press of Virginia, Charlottesville.
691 p.
Phillips, W.
Pond,
S.
1963.
Depth of roots in soil.
Ecology 44:424.
W., and D. R. Cable.
1960.
Effect of heat treatment on sprout production of
some shrubs of the chaparral in central Arizona.
J. Range Manage.
13:313 -317.
F.
Rowe, P. B., and L.
vegetation.
J.
Reimann.
1961.
For. 59:175 -181.
R.
Saunier, R. E., and R. F. Wagle.
Ecology 46:749 -750.
Water use by brush, grass, and grass -forb
1965.
Root grafting in Quercus turbinella Greene.
Saunier, R. E., and R. F. Wagle.
1967.
Factors affecting the distribution of shrub
live oak.
(Quercus turbinella Greene).
Ecology 48:35 -41.
Weaver, J.
E., and F. C. Clements.
1938.
Plant Ecology.
248
McGraw -Hill, New York. 601 p.